Mitigating Hallucinations in Large Vision-Language Models (LVLMs) via Language-Contrastive Decoding (LCD)

Avshalom Manevich, Reut Tsarfaty


Abstract
Large Vision-Language Models (LVLMs) are an extension of Large Language Models (LLMs) that facilitate processing both image and text inputs, expanding AI capabilities. However, LVLMs struggle with object hallucinations due to their reliance on text cues and learned object co-occurrence biases. While most research quantifies these hallucinations, mitigation strategies are still lacking. Our study introduces a Language Contrastive Decoding (LCD) algorithm that adjusts LVLM outputs based on LLM distribution confidence levels, effectively reducing object hallucinations. We demonstrate the advantages of LCD in leading LVLMs, showing up to %4 improvement in POPE F1 scores and up to %36 reduction in CHAIR scores on the COCO validation set, while also improving captioning quality scores. Our method effectively improves LVLMs without needing complex post-processing or retraining, and is easily applicable to different models. Our findings highlight the potential of further exploration of LVLM-specific decoding algorithms.
Anthology ID:
2024.findings-acl.359
Volume:
Findings of the Association for Computational Linguistics: ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6008–6022
Language:
URL:
https://aclanthology.org/2024.findings-acl.359
DOI:
10.18653/v1/2024.findings-acl.359
Bibkey:
Cite (ACL):
Avshalom Manevich and Reut Tsarfaty. 2024. Mitigating Hallucinations in Large Vision-Language Models (LVLMs) via Language-Contrastive Decoding (LCD). In Findings of the Association for Computational Linguistics: ACL 2024, pages 6008–6022, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Mitigating Hallucinations in Large Vision-Language Models (LVLMs) via Language-Contrastive Decoding (LCD) (Manevich & Tsarfaty, Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-acl.359.pdf